Prevalence of obesity according to body mass index, waist circumference, and waist-to-height ratio in Peru: A systematic review and meta-analysis

Abstract

Introduction

Obesity is a global public health epidemic with significant health implications. However, studies on the prevalence of obesity in Peru have yielded varied results, highlighting the need for updated data to inform effective public health policies.

Objective

The primary objective of this study is to determine the prevalence of obesity in Peru using three anthropometric measures: body mass index (BMI), waist circumference (WC), and waist-to-height ratio (WHtR).

Methods

Between March and April 2024, a systematic review of published studies reporting the prevalence of obesity in Peru was conducted. The databases Scopus, Web of Science, Embase, PubMed, LILACS, and Scielo were Searched.

Results

Overall, the prevalence of obesity was 23.23 %, 38.90 %, and 81.53 % according to BMI, WC, and WHtR, respectively. However, these figures show wide variability, ranging from 13.10 % to 37.4 % according to BMI and from 19.4 % to 51.6 % according to WC. The highest reported prevalence of obesity by WHtR was 85.4 %. Nonetheless, only a fraction of these studies were published in the last five years, and few specifically focused on obesity as the primary objective.

Conclusions

The prevalence of obesity in Peru varies significantly depending on the anthropometric measure used. To improve the collection and frequency of data on obesity in Peru, it is recommended that cut-off points be standardized to be suitable for the country and that annual national surveys specifically designed for this purpose be implemented.

Graphical abstract

1. Introduction

Obesity is among the most urgent health epidemics worldwide. Obesity directly and indirectly contributes to metabolic diseases, increases mortality rates, and leads to numerous health complications [1]. Projections say by 2025, one in five adults will face obesity globally. Regions like Southeast Asia and Africa are expected to see big jumps in these numbers. Economically, this epidemic accounts for about 13 % of all healthcare costs—almost $990 million yearly [2].

In Latin America, roughly a quarter of people struggle with obesity; the trend keeps climbing despite global changes. Data from the Latin American Study of Nutrition and Health (ELANS, an acronym in Spanish), conducted between 2014 and 2015, shows that of 9218 participants, around 25.2 % were obese. In Peru specifically, this study found an obesity rate of 22 %, using BMI as the measure [3].

Obesity rates in Peru vary according to national surveys. The National Household Survey (ENAHO) from 2012 to 2013 indicated an obesity prevalence of 19.7 % using BMI [4]. Additionally, the Food and Nutrition Surveillance by Life Stage Adult Survey (VIANEV, acronym in Spanish) from 2017 to 2018 showed higher numbers: 26.8 % with BMI, 50.4 % with waist circumference (WC), and up to 85.4 % with waist-to-height ratio (WHtR) [5]. A more recent study by Vásquez-Romero utilizing data from the National Demographic and Health Survey (ENDES, Spanish acronym) showed national obesity rates of 25.65 %, 42.04 %, and 46.49 %, based on BMI, WC, and WHtR, respectively [6].

These differing results highlight the need to investigate how obesity prevalence differs according to different measures at a national level; hence, this study aims to find out how common obesity is in Peru by looking at three measures—BMI, WHtR, and WC—to give a comprehensive perspective of the epidemic within Peru context.

2. Methodology

2.1. Study design

This study is a systematic review and meta-analysis of the prevalence of obesity in Peru. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 statement was used to ensure the proper conduct of systematic reviews and meta-analyses [7]. Our analysis complied with all 27 items in the PRISMA guidance checklist. The main adaptation in our methodology was the inclusion of non-indexed national data sources to complement scientific publications, specifically the ENINBSC database and PERU MIGRANT study, which were essential for providing a comprehensive view of obesity prevalence in Peru. This adaptation was necessary due to the limited availability of indexed publications meeting all our objectives.

2.2. Search strategy

Between March 1 and April 1, 2024, a systematic search was conducted in Scopus, Web of Science, Embase, PubMed, Latin American and Caribbean Health Sciences Literature (LILACS), and Scientific Electronic Library Online (SciELO) databases. The key terms used included "prevalence," "obesity," "body mass index," "abdominal circumference," and "waist-to-height ratio." The complete search strategy is available in Supplementary Material 1.

2.3. Selection criteria

The inclusion of studies was evaluated based on the following criteria: 1) original articles, 2) those employing probabilistic sampling techniques, which are essential to ensure that the results are generalizable to the study population, and 3) those that assessed obesity using at least one of the following anthropometric parameters: BMI, WC, or WHtR. Conversely, studies focused on selective populations, such as groups with specific health conditions, were excluded as these could introduce biases in evaluating obesity prevalence.

2.4. Definition of obesity measures

This systematic review analyzed obesity using three anthropometric measures and their respective cut-off points. For BMI, obesity was defined as BMI ≥30 kg/m², a standardized criterion across all included studies.

The literature identified two main criteria for WC: the National Cholesterol Education Program ATP III criteria (men: ≥102 cm; women: ≥88 cm) and the International Diabetes Federation (IDF) criteria (men: ≥90 cm; women: ≥80 cm). To ensure methodological consistency, only studies using ATP III criteria were included for meta-analysis purposes.

For the WHtR, while some studies reported alternative cut-off points (e.g., ≥0.59), we used the standard cut-off point of ≥0.50. Only studies using this standard ≥0.50 cut-off were included to maintain consistency in our meta-analysis.

2.5. Study selection

With the search results, the first phase involved the removal of duplicates using the Rayyan software for article storage. Abstracts of articles/conferences and unavailable manuscripts were excluded.

In the second phase, two investigators (FEZM and JLC) independently reviewed the titles and abstracts of the manuscripts. If there was agreement on inclusion, the manuscript was added; otherwise, it was excluded. In case of any discrepancy or doubt about inclusion, a third author (LEMVR) resolved it.

Finally, the full text of the included manuscripts was verified and added to an Excel sheet. The investigators also carried out this process, and the fourth author resolved any doubts.

2.6. Data Extraction and Qualitative analysis

Using Microsoft Excel 2016, the following specific data were extracted from each study: first author and year, city, inclusion criteria, exclusion criteria, sample size, sex (% female), age, and diagnosis of obesity according to BMI, WC, and WHtR.

2.7. Risk of bias assessment

Munn Z et al. [8] assessed the quality of the selected articles using the risk of bias tool for prevalence studies. This tool consists of nine questions with three standard response options (include/exclude/seek more information) based on the evaluator's criteria.

The study's risk of bias was evaluated based on the following nine criteria: 1) Was the sample frame appropriate to address the target population? 2) Were the study participants recruited appropriately? 3) Was the sample size adequate? 4) Were the study subjects and setting described in detail? 5) Was the data analysis conducted with sufficient coverage of the identified sample? 6) Were valid methods used for the identification of the condition? 7) Was the condition measured in a standard and reliable manner for all participants? 8) Was the statistical analysis appropriate? 9) Was the response rate adequate, and if not, was the low response rate managed appropriately?

The level of bias is assessed by calculating the total number of criteria with an affirmative response and converting this score into a percentage (n/9). Studies scoring below 50 % are considered high risk of bias, 50–69 % moderate risk of bias, and 70 % and above low risk of bias. The quality assessment tool was initially tested on a small number of studies. Two additional investigators were available for consultations and to resolve any disagreements.

2.8. Statistical analysis

All quantitative analyses were performed using STATA version 18 (Stata Corp, College Station, TX, USA). All sources providing data on BMI, WC, and WHtR prevalence were included in the meta-analysis and pooled analysis. The total sample size and the number of cases for each condition were extracted to calculate prevalences and their 95 % confidence intervals (95 % CI).

A random-effects model was used to combine the prevalences, as significant heterogeneity was expected between the studies due to differences in study populations, measurement methods, and other factors. Heterogeneity among the studies was assessed using the I² statistic. The interpretation of heterogeneity followed the Cochrane Handbook guidelines: 0–40 % = might not be important; 30–60 % = moderate heterogeneity; 50–90 % = substantial heterogeneity; 75–100 % = considerable heterogeneity.

Sensitivity analyses were performed to assess the robustness of the results by excluding studies with a high risk of bias or different measurement methods. Additionally, sensitivity analyses were conducted considering three key aspects: temporal trends (differentiating between <2013 and ≥ 2013), sex distribution, and geographical scope (subnational vs. national). The year 2013 was selected as a temporal cut-off point due to significant changes in Peru's public health landscape during this period, including the standardization of anthropometric measurement protocols in national surveys and improvements in data collection methodologies. This temporal division allows us to evaluate potential changes in obesity prevalence while considering these methodological improvements in data collection and reporting.

2.9. Ethical considerations

This study is based on analyzing manuscripts published in journals and scientific databases, so the risks involved are minimal. Therefore, the authors did not consider it necessary to submit the project to a research ethics committee.

3. Results

3.1. Study selection

A total of 4529 publications were found, and after removing duplicates, 3094 unique articles remained. From these, 1435 manuscripts were evaluated, of which 70 articles met the inclusion criteria. Finally, 17 manuscripts were included in the analysis. Details are provided in Fig. 1.

Fig. 1.

Flow chart.

3.2. Study characteristics

The studies, ranging in date from Jacoby (2003) [9] to Vásquez-Romero (2024) [6], presented sample sizes from 217 [10] to 35 255 individuals [6]. Four studies were national in scope, while 13 were subnational. Of the total, 16 studies evaluated obesity by BMI, 12 by WC, and 5 by WHtR (Table 1).

Table 1.

Summary of studies on the prevalence of obesity (BMI, WC, WHR) in Peru.

First author, year	City	Inclusion Criteria	Exclusion Criteria	Sample	Gender (% women)	Age (mean, median, or proportion)	Obesity measure evaluated	Diagnostic criteria	Database (year of collection)
Jacoby, 2003	Lima, Huánuco, Arequipa, Ucayali	Those who wanted to be part of the study	Those who did not have the cognitive ability to respond to the study	2322	1157 (49.8 %)	H: 42.8 M: 38.2	BMI WHtR	≥30 m/Kg2 men ≥94 cm women ≥84 cm	JACOBY
ENINBSC 2004–2005 MINSA	Nacional	Over 20 years old, oriented in time, space, and place, aware of not having a serious illness	Pregnant, breastfeeding, those who have altered their usual diet due to illness, party and/or celebrations, people with physical disabilities that do not allow normal evaluation of anthropometric indicators, as well as bodybuilders and qualified athletes	4183	2097 (50.13 %)	20–38	BMI WC WHtR	≥30 m/Kg2 men ≥102 cm women ≥88 cm ≥0.50	ENINBSC (2004–2005)
Seclen, 2006	Rímac, SMP, Los Olivos	Over 30 years old and residents in the mentioned areas	Type 1 diabetes, pregnant, acute illness, presence of a medical condition that could affect anthropometric measures or laboratory data, illiterate	838	612 (68.3 %)		BMI WC	≥30 m/Kg2 men ≥102 cm women ≥88 cm	Propia
Medina-Lezama, 2007	Arequipa	Between 20 and 80 years old	Pregnant	1878	1011 (53.83 %)	49.6 48.5	BMI	≥30 m/Kg2	PREVENTION (2005)
Schargrodsky, 2008	Lima	Over 25 years old up to 64 years old	Life limits, pregnant, residents with an address different from their home or residents who lived in areas considered dangerous for interviewers and people who were visiting	1652	883 (53.4 %)	43.6	BMI	≥30 m/Kg2	CARMELA (2008)
Miranda JJ, 2011	Lima - Ayacucho	Over 30 years old	Those with incomplete laboratories or anthropometric measurements	987	521 (52.8 %)	48	BMI WC WHtR	≥30 m/Kg2 men ≥102 cm women ≥88 cm ≥0.50	PERU MIGRANT (2007–2008)
Revilla L., 2014	Lima, Callao	Over 15 years old	Under 15 years old, residence less than a year in the study location, pregnant, and disabling conditions that prevent adequate anthropometric evaluation	1771	1105 (62 %)	39.5	BMI WC	≥30 m/Kg2 men ≥102 cm women ≥88 cm	FRENT LIMA CALLAO (2007)
Herrera-Enriquez, 2017	Arequipa - Chivay	Individuals over 40 years of age	Individuals with a history or evidence of alcoholism, advanced liver damage with ascites, renal insufficiency, hypothyroidism, Cushing's syndrome, psychiatric illness, pregnant or postpartum women, use of oral contraceptives, hormone replacement therapy, steroids, glucocorticoids, and incomplete clinical history data	237	127 (53.6 %)	H: 68.9 M: 62.0	BMI WC	≥30 m/Kg2 men ≥102 cm women ≥88 cm	Propia
Adams, 2018	Lima	Individuals aged 20–59 years, beneficiaries of popular dining halls in Lima	Individuals with terminal chronic disease, mental deficiency or difficulty in expression, consumptive disease, pregnant women, or those with acute illness	374	266 (71.1 %)		BMI WC	≥30 m/Kg2 men ≥90 cm women ≥80 cm	Propia
Bernabé-Ortiz, 2018	Tumbes	Individuals aged 30–69 years, full-time residents in the study area (≥6 months) and capable of understanding the procedures and providing informed consent	Pregnant women or those with any physical disability that impeded anthropometric measurements (weight, height, blood pressure, or waist circumference) or bedridden individuals. Participants who did not complete laboratory analysis	1612	810 (50.3 %)	48.2	BMI WC	≥30 m/Kg2 men ≥90 cm women ≥80 cm	FINDRISCK (2014)
Carrillo-Larco, 2018	Lima, Tumbes, Puno rural y urbano	Individuals over 35 years of age	Pregnant women who could not provide informed consent, individuals incapable of responding to the questionnaires	3217	1660 (51.6 %)	55.7	BMI WC	≥30 m/Kg2 men ≥90 cm women ≥80 cm	CRONICAS (2010–2011)
Paz-Krumdiek, 2019	Nacional	Individuals over 30 years of age	Extremes of life, pregnant women, postpartum women, and those not presenting variables of interest	8587	4602 (53.6 %)	38.4	BMI	≥30 m/Kg2	ENAHO-CENAN (2011)
Barboza Palomino, 2020	Huamanga (Ayacucho)	Personas entre 18 a 64 años que viven en ayacucho	Pregnant or breastfeeding women, individuals with neurological or mental health issues that impede a logical interview or responses	412			BMI	≥30 m/Kg2	Propia
Bernabé-Ortiz, 2020	Tumbes	Individuals over 18 years of age, capable of understanding the procedures, able to provide informed consent, and full-time residents in the area	Individuals with self-reported history of chronic kidney disease and heart disease under treatment with digoxin	2376	1197 (50.4 %)	43.3	BMI	≥30 m/Kg2	SALT (2014)
Aparco, 2022	Nacional	Individuals aged 18–59 years, fasting for no less than 9 h and no more than 12 h	Adults aged 18–59 years not on the identification list, pregnant/postpartum women, adults under treatment that alters glucose or lipid profile, those not fasting, individuals with gastrointestinal issues that alter food intake, and genetic conditions or malformations limiting anthropometric techniques	1047	603 (57.6 %)	21.4	BMI WC WHtR	≥30 m/Kg2 men ≥94 cm women ≥88 cm ≥0.50	VIANEV (2017–2018)
Vasquez Romero, 2024	Nacional	Individuals aged 20–99 years	Pregnant women, absent anthropometric markers, or values outside the normal physiological range	35 255	18 136 (51.44)	40.78	BMI WC WHtR	≥30 m/Kg2 men ≥102 cm women ≥88 cm ≥0.50	ENDES 2022

A variety of approaches were observed regarding the databases used. Some studies developed and used databases tailored to their specific research needs. For example, studies by Seclen [11], Herrera-Enriquez [10], Adams and Chirinos [12], and Barboza Palomino [13] designed their datasets to study obesity in specific populations in Lima, Arequipa, and Huancayo. Conversely, several studies utilized existing databases from previous research, offering an established framework for longitudinal or comparative analysis. The National Survey of Indicators for Non-Communicable Chronic Diseases (ENINBSC) [14] collected diet, exercise, and other anthropometric indicators data. Medina-Lezama [15] used data from the PREVENTION study, while Revilla L [16]. employed data from the FRENT LIMA-CALLAO study, and Carrillo-Larco et al. [17] used data from the CRONICAS study. Additionally, some studies used national survey databases such as ENAHO [4], VIANEV [5], and ENDES, which provide extensive and representative information on the Peruvian population. These surveys are particularly valuable for epidemiological studies as they include a wide range of data on health, nutrition, and socioeconomic factors at the national level.

Significant variations were observed in the analysis of obesity prevalence across different studies. The study by Adams and Chirinos reported the highest prevalence for both BMI (37.40 %) and WC (51.6 %) [12]. In contrast, the ENINBSC study [14] showed the lowest prevalence for BMI at 13.10 %, and Jacoby [9] recorded the lowest for WC at 19.4 % [9]. Regarding WHtR, Aparco and Cárdenas-Quintana [5] reported the highest prevalence at 85.40 %, while Vásquez-Romero [6] reported two prevalences for WHtR depending on the cut-off point: ≥0.50 (84.78 %) and ≥0.59 (48.05 %).

The percentage of female participants in the reviewed studies varied considerably, with the lowest being 50.13 % (Adams and Chirinos) [12] and the highest at 71.1 % (ENINBSC) [14], which could influence the prevalence variations observed between studies. Regarding the geographical distribution of the studies, there was a higher concentration of research in urban areas, particularly in the capital, Lima, followed by major cities such as Tumbes and Arequipa.

Among studies that included younger populations, Revilla L [16]. included individuals aged 15, standing out with the lowest age limit among all listed. On the other hand, studies with a broader age range, with an upper limit of 99 years, include those by Vásquez-Romero [6], ENINBSC [14], and Seclen [11], covering a considerable diversity in adult and elderly ages.

During this systematic review, a limitation was encountered in not identifying manuscripts published in indexed databases that met all proposed objectives. However, aware of the importance of a comprehensive approach, we resorted to non-indexed national data sources to complement scientific publications. Specifically, we accessed and extracted pertinent data from ENINBSC [14], which, despite not being indexed, provides vital information for our research. Similarly, we analyzed the PERU MIGRANT study database [18], using the study by Miranda JJ [19]. as a reference. Although the associated publication did not report obesity prevalence data by WC and WHtR, we extracted these data directly from their database. This strategy allowed us to include a broader range of data and provide a more robust and representative analysis of obesity prevalence in Peru, thereby maximizing the utility and impact of our findings in understanding this public health epidemic.

3.3. Risk of bias assessment

The 19 selected articles were evaluated using the risk of bias scale by Munn et al. [8] for prevalence studies (Table 2). The studies being assessed showed adequate compliance with criteria to minimize the risk of bias, using appropriate sampling frameworks and effective recruitment methods to select representative samples of the target population. Additionally, all studies confirmed adequate sample sizes, provided detailed descriptions of subjects and settings, and conducted thorough data analyses, ensuring the validity and reliability of their results.

Table 2.

Level of bias in studies on the prevalence of obesity in peru.

Estudio, año	1	2	3	4	5	6	7	8	9
Study, Year	Was the sampling frame appropriate to address the target population?	Were the study participants recruited in an appropriate manner?	Was the sample size adequate?	Were the subjects and the setting of the study described in detail?	Was the data analysis conducted with sufficient coverage of the identified sample?	Were valid methods used for the identification of the condition?	Was the disease measured in a standard and reliable manner for all participants?	Was the statistical analysis appropriate?	Was the response rate adequate, and if not, was the low response rate handled appropriately?	Final Score
Jacoby, 2003	∗	∗	∗	∗	∗	∗	∗	∗		8	Low
ENINBSC 2004–2005 MINSA	∗	∗	∗	∗	∗	∗	∗	∗	∗	9	Low
Seclen, 2006	∗	∗	∗	∗	∗	∗	∗	∗		8	Low
Medina-Lezama, 2007	∗	∗	∗	∗	∗	∗	∗	∗	∗	9	Low
Schargrodsky, 2008	∗	∗	∗	∗	∗	∗	∗	∗	∗	9	Low
Miranda JJ, 2011	∗	∗	∗	∗	∗	∗	∗	∗	∗	9	Low
Revilla L., 2014	∗	∗	∗	∗	∗	∗	∗	∗		8	Low
Herrera-Enriquez, 2017	∗	∗	∗	∗	∗	∗	∗	∗		8	Low
Adams y Chirinos, 2018	∗	∗	∗	∗	∗	∗	∗	∗		8	Low
Bernabé-Ortiz, 2018	∗	∗	∗	∗	∗	∗	∗	∗		8	Low
Carrillo-Larco, 2018	∗	∗	∗	∗	∗	∗	∗	∗	∗	9	Low
Paz-Krumdiek, 2019	∗	∗	∗	∗	∗	∗	∗	∗	∗	9	Low
Barboza Palomino, 2020	∗	∗	∗	∗	∗	∗	∗	∗		8	Low
Bernabé-Ortiz, 2020	∗	∗	∗	∗	∗	∗	∗	∗	∗	8	Low
Aparco y Cárdenas-Quintana, 2022	∗	∗	∗	∗	∗	∗	∗	∗		8	Low
Vasquez Romero, 2024	∗	∗	∗	∗	∗	∗	∗	∗	∗	8	Low

However, the response rate and management were the most variable points. Although not all studies provided explicit details on how a potentially low response rate was handled, those that did maintained a low risk of bias.

3.4. Meta-analysis on obesity prevalence

According to BMI, the pooled prevalence of obesity was 23.23 % (95 % CI: 20.00%–26.63 %), but the heterogeneity was quite high (I² = 99.2 %) (Fig. 2).

Fig. 2.

Pooled prevalence of obesity according to BMI in Peru.

The systematic review results indicate significant variations in obesity prevalence measured by BMI across different categories. Ten studies were included for each sex, finding an obesity prevalence in males of 12.74 % (95 % CI: 5.93 %–23.22 %, I² = 99.09 %), while in females, the prevalence was considerably higher at 42.9 % (95 % CI: 31.33 %–54.89 %, I² = 99.23 %). Analyzing the data over time, with two studies in each category, an increase in obesity prevalence was observed in studies conducted from 2013 onwards (43.11 %, 95 % CI: 42.61 %–43.62 %) compared to those conducted before 2013 (29.5 %, 95 % CI: 28.26 %–30.75 %). Regarding geographical scope, the prevalence in subnational studies was 31.31 % (95 % CI: 29.59 %–33.05 %), while national studies showed a slightly lower prevalence of 27.77 % (95 % CI: 27.33 %–28.21 %) (Table 3).

Table 3.

Sensitivity analysis of obesity prevalence by BMI in Peru.

Category	Number of studies	Prevalence	IC 95 %	I2
Sex
Men	10	16.86	13.09, 20.99	97.34
Women	10	27.15	22.81, 31.72	97.56
Time
<2013	5	19.13	15.04, 23.60	96.4
≥2013	10	25.40	21.64, 29.36	98.46
Scope
Subnational	11	24.35	21.78, 27.03	92.89
National	4	20.17	12.88, 28.61	99.66

For obesity, according to WC, only those studies that used cut-off points according to the National Cholesterol Education Program Adult Treatment Panel III (ATP III) were included in the meta-analysis. In this way, the pooled prevalence was 38.90 % (95 % CI: 37.47 %–46.60 %), but the heterogeneity was quite high (I² = 99.02 %) (Fig. 3).

Fig. 3.

Pooled prevalence of obesity according to WC in Peru.

In the sensitivity analysis of obesity prevalence by WC in Peru, ten studies were evaluated for each sex group. The prevalence of obesity in males was 17.10 %, with a 95 % confidence interval ranging from 10.03 % to 25.58 %, showing very high heterogeneity (I² = 98.81 %). In contrast, the prevalence in females was significantly higher, estimated at 49.58 % (95 % CI: 34.87 %–64.32 %), with also high heterogeneity (I² = 99.50 %). Over time, studies conducted before 2013 reported a prevalence of 33.64 % (95 % CI: 32.36 %–34.93 %), whereas more recent studies conducted from 2013 onwards showed an increase in prevalence to 43.11 % (95 % CI: 42.61 %–43.62 %). From a geographical perspective, subnational studies indicated a prevalence of 39.28 % (95 % CI: 37.46 %–41.11 %), compared to national studies, which reported a lower prevalence of 27.43 % (95 % CI: 27.33 %–28.21 %) (Table 4).

Table 4.

Sensitivity analysis of obesity prevalence by WC in Peru.

	Number of studies	Prevalence	IC 95 %	I2
Sex
Men	10	12.74	5.93, 23.22	99.09
Women	10	42.9	31.33, 54.89	99.23
Time
<2013	2	29.5	28.26, 30.75
≥2013	2	43.11	42.61, 43.62
Scope
Subnational	2	31.31	29.59, 33.05
National	2	27.77	27.33, 28.21

According to WHtR, the pooled prevalence for obesity was 81.53 % (95 % CI: 77.56%–85.19 %), but the heterogeneity was quite high (I² = 98.61 %) (see Fig. 4).

Fig. 4.

Pooled prevalence of obesity according to WHtR in Peru.

The sensitivity analysis for obesity prevalence determined by WHtR in Peru included four studies for each sex group. Male prevalence was 76.87 % (95 % CI: 72.33 %–81.11 %), with a heterogeneity of 94.72 %. In females, the prevalence was slightly higher, at 87.58 % (95 % CI: 84.38 %–90.45 %), with a heterogeneity of 93.74 %. Regarding the time of the studies, those conducted before 2013 presented a prevalence of 76.76 % (95 % CI: 78.65 %–80.84 %), while more recent studies, from 2013 onwards, showed a prevalence of 82.91 % (95 % CI: 77.33 %–87.85 %), with extremely high heterogeneity of 99.15 %. Analyzing the geographical scope, subnational studies revealed a prevalence of 82.19 % (95 % CI: 77.79 %–86.21 %), compared to national studies, which showed a similar prevalence of 76.87 % (95 % CI: 72.33 %–81.11 %). Heterogeneity was also very high in both cases, at 98.9 % for subnational and 94.72 % for national (Table 5).

Table 5.

Sensitivity analysis of obesity prevalence by WHtR in Peru.

	Number of studies	Prevalence	IC 95 %	I2
Sex
Men
Women	4	76.87	72.33, 81.11	94.72
Time	4	87.58	84.38, 90.45	93.74
<2013
≥2013	2	76.76	78.65, 80.84
Scope	3	82.91	77.33, 87.85	99.15
Subnational
National	1	78.62	75.96, 81.07

4. Discussion

This systematic review was designed to assess the prevalence of obesity in Peru using anthropometric parameters such as BMI, WC, and WHtR. Including 16 studies that met rigorous selection criteria, the analysis provides a crucial perspective on the evolution and distribution of obesity across various demographics and regions in the country. However, it is concerning that only five of these studies were published in the last five years [[4], [5], [6],13,20], and only two studies [5,6] aimed to determine obesity prevalence. This indicates a possible lack of recent data that could affect the responsiveness of public health policies to this escalating epidemic.

The results reveal an upward trend in obesity prevalence since 2013, an alarming signal that could be related to lifestyle changes, such as increased physical inactivity and the consumption of processed and high-energy-density foods [21]. Specifically, the gender disparity is notable, with women showing a significantly higher prevalence compared to men. This difference could reflect variations in biological, behavioral, and socioeconomic factors that disproportionately affect women, highlighting the need for public health strategies that consider the particularities of each sex [22].

The markers used to measure obesity showed considerable variations in their results. WC and WHtR showed prevalences that suggest conventional cut-off points may not be appropriate for the Peruvian population. For example, the ATP III criterion [23] for WC was preferred over the International Diabetes Federation (IDF) criterion [24], as the latter estimated a prevalence of 60 %, an unrealistic figure for the population under study. Similarly, for WHtR, if the globally consensual cut-off points of ≥0.5 [25] are followed, obesity prevalence would reach up to 80 %, a value that, as mentioned earlier, does not reflect reality.

The markers used to measure obesity showed considerable variations in their results, revealing significant disparities depending on the criteria used. Our findings demonstrate that obesity prevalence varies substantially: 23.23 % when using BMI, 38.90 % using WC with ATP III criteria (≥102 cm for men and ≥88 cm for women), and increasing dramatically to approximately 60 % when applying IDF criteria (≥90 cm for men and ≥80 cm for women). This striking variation raises important methodological considerations, particularly since these cut-off points were not developed or validated in Peruvian or Latin American populations.

The markers used to measure obesity showed considerable variations in their results, revealing significant disparities depending on the criteria used. Our findings demonstrate that obesity prevalence varies substantially: 23.23 % when using BMI, 38.90 % using WC with ATP III criteria (≥102 cm for men and ≥88 cm for women), and increasing dramatically to approximately 60 % when applying IDF criteria (≥90 cm for men and ≥80 cm for women). This striking variation raises important methodological considerations, particularly since these cut-off points were not developed or validated in Peruvian or Latin American populations. Most notably, the WHtR measurements using the standard cut-off point of ≥0.5 resulted in an exceptionally high prevalence of 81.53 %, which raises particular methodological concerns. This markedly high WHtR prevalence, compared to other anthropometric measures, strongly suggests that the current globally recommended WHtR threshold may require specific validation and potential recalibration for the Peruvian population to ensure it accurately reflects obesity-related health risks.

The ATP III criteria were primarily established from studies in Caucasian populations [26], while the IDF criteria for Latin America were largely extrapolated from Asian population studies. This raises questions about their applicability to the Peruvian population. Similar concerns have been raised in other Latin American countries [[27], [28], [29]]; for instance, studies in Latin American countries have found that international cut-off points may not accurately reflect adiposity-related cardiometabolic risk in their populations.

The relationship between waist circumference and ethnicity is particularly relevant for Latin American populations with distinct anthropometric characteristics. While the IDF provides specific WC cut-off points for Central and South American populations (≥90 cm for men and ≥80 cm for women), these were primarily extrapolated from Asian data rather than being derived from Latin American studies [30]. No unified database or consensus is specifically validated for Latino/Latina populations. Recent systematic reviews have highlighted this gap, noting significant variations in body composition and fat distribution patterns among Latin American countries [31]. This lack of region-specific validated cut-off points remains a considerable challenge for obesity assessment in Latin American populations, including Peru, highlighting the need for large-scale validation studies.

The situation is particularly concerning for WHtR, where the globally recommended cut-off point of ≥0.5 results in an obesity prevalence of 81.53 %, a figure substantially higher than those obtained with other anthropometric measures. This marked disparity raises serious methodological concerns about the appropriateness of current WHtR thresholds for the Peruvian population. While WHtR has been promoted as a simple screening tool for obesity and cardiometabolic risk, our findings suggest that its current cut-off point may significantly overestimate obesity prevalence in Peru. This disparity between different anthropometric measures strongly emphasizes the need for population-specific validation studies. Recent research has noted that anthropometric cut-off points should consider ethnic variations in body composition and fat distribution patterns [32]. Therefore, while these international criteria provide a starting point for obesity assessment, our findings suggest they may need to be recalibrated for the Peruvian population to provide more accurate estimates of obesity-related health risks, with particular attention to establishing appropriate WHtR thresholds through rigorous validation studies.

Additionally, the period of our included studies (2003–2024) reflects our analysis's strengths and limitations. While this wide range allows us to observe temporal trends in obesity prevalence, it may also contribute to the observed variability in our results. The increasing trends we observed in our temporal sensitivity analyses align with global patterns of rising obesity rates. However, we note that most of our included studies (12 out of 16) were conducted after 2013, providing greater weight to more recent data in our pooled estimates. The observed increases in prevalence across all anthropometric measures between the pre-2013 and post-2013 periods suggest that temporal changes in obesity rates are a significant factor contributing to the variability in our findings.

Another methodological consideration is the potential variability and reliability of WC measurement techniques. Among our included studies, there were variations in measurement protocols: some measured WC at the midpoint between the lowest rib and iliac crest, others at the umbilical level, and some at the superior border of the iliac crest. This lack of standardization could contribute to the observed variations in prevalence. Furthermore, none of the studies reported inter-observer reliability data for WC measurements or addressed the known challenges in obtaining consistent measurements, particularly in individuals with higher degrees of adiposity. These methodological inconsistencies, combined with the inherent difficulties in WC measurement (such as proper positioning of the measuring tape, variations in breathing patterns, and anatomical changes in individuals with central obesity), could partly explain the discrepancies observed in our analysis.

4.1. Challenges and limitations of current national surveys

Only four of the 16 studies identified in our systematic review are national in scope, highlighting a significant limitation in the coverage and periodicity of obesity data in Peru. The ENDES survey is conducted annually [33] but is not specifically designed to measure obesity. Although it includes a section on obesity, its focus is not as comprehensive as a survey dedicated exclusively to this epidemic. Similarly, the ENAHO [34] occasionally reports obesity prevalence solely based on BMI, which does not provide a complete picture.

On the other hand, the ENINBSC [14] and VIANEV [35] surveys, although more specific, are outdated; the former was conducted in 2005, and the latter in 2017, and they are not updated annually. This contrasts markedly with surveys like Mexico's ENSANUT [36], targeted at metabolic problems and updated regularly, providing continuous and up-to-date data crucial for formulating effective public health policies.

This review highlights the urgency of implementing annual national surveys to monitor obesity in all dimensions, including BMI and WC. Greater investment in such studies would allow for more accurate and representative data on the prevalence of obesity in Peru, facilitate monitoring trends over time, and assess the effectiveness of implemented interventions.

Establishing annual and specific surveys for obesity would help align Peru with international standards and enable a more agile and informed response to this growing public health challenge. These studies should include various anthropometric markers to provide a more comprehensive view of the population's nutritional and health status. Additionally, they should be designed to capture significant regional and demographic differences, allowing for more targeted and effective interventions.

The Peruvian government and public health agencies must recognize the importance of these data and provide the necessary resources for their systematic collection and analysis. Updating and expanding the data infrastructure on obesity would facilitate a better understanding of the epidemic, the adaptation of public policies to changing realities, and the implementation of more effective prevention and treatment strategies based on solid and up-to-date evidence. This is supported by successful examples from other countries, such as Mexico's National Health and Nutrition Survey, demonstrating how regular, comprehensive health surveillance can inform effective policy interventions [37]. Similarly, Chile's National Health Survey has shown how robust data infrastructure can support the development and evaluation of public health policies, such as their pioneering food labeling law [38]. These experiences highlight how strengthened surveillance systems can lead to more targeted and effective obesity prevention strategies.

4.2. Importance of the study for public health

The importance of this study for public health in Peru is significant, given the growing obesity epidemic and its associated consequences. Obesity is not just an individual health problem but a public health challenge that incurs high costs for the healthcare system, reduces quality of life, and increases mortality. By providing an updated assessment of obesity prevalence using various anthropometric markers, this study offers a crucial foundation for understanding the scale of the problem and identifying the population groups at higher risk. This information is vital for designing and implementing effective and targeted public health policies and intervention programs.

Moreover, the study emphasizes the need to improve data collection and the frequency of health surveys related to obesity in Peru. The lack of recent and specific data on obesity limits the ability of the government and health organizations to develop policies that effectively address this crisis. With more frequent and particular studies, such as annual and nationally representative surveys, policymakers would have access to information that more accurately reflects current trends and the effectiveness of ongoing interventions. This is essential for assessing the progression of obesity at the national level and adjusting intervention strategies as new evidence emerges.

Finally, this study highlights the importance of adjusting the cut-off points and methodologies used to measure obesity in the Peruvian context, which could result in more accurate estimates and, consequently, more appropriate interventions. Adapting international standards to local characteristics is crucial for accurately assessing the population's health and developing health programs that address Peru's cultural, economic, and social particularities. Ultimately, conducting such studies strengthens the country's ability to manage public health more effectively, offering hope for a significant reduction in the burden of obesity for future generations.

4.3. Study limitations

A crucial limitation of this study is the limited availability of recent research specifically focused on obesity. Another significant limitation is the variability in methods and cut-off points used to define obesity among the different studies included in the review. Differences in how obesity is defined and measured, such as the cut-off points for WC and WHtR, can lead to discrepancies in reported prevalences. This lack of standardization complicates direct comparisons between studies and may affect the accuracy of national prevalence estimates, hindering the formulation of effective intervention strategies.

Moreover, although some included studies are national in scope, many are subnational and may not fully capture regional variations within Peru. This limitation is relevant because it prevents identifying specific areas or population groups that require urgent interventions. Additionally, the infrequency of national studies reduces the ability to continuously assess the impact of public health interventions and adjust policies according to emerging needs.

Finally, another significant limitation is the WHtR measurements, which showed particularly high prevalence rates (81.53 %) using current cut-off points. This finding highlights a critical methodological challenge: the standard WHtR threshold of ≥0.5 was primarily validated in European and Asian populations, with limited validation studies in Latin American contexts. The markedly high prevalence obtained with this measure suggests that this universal cut-off point may not be appropriate for the Peruvian population. This limitation is especially important given that WHtR has been promoted as a simple screening tool for obesity and cardiometabolic risk. Future research should prioritize validation studies to establish population-specific WHtR thresholds for Peru, ideally correlating different cut-off points with cardiometabolic outcomes in this population.

5. Conclusions

• •This study has provided a comprehensive view of the prevalence of obesity in Peru using a variety of anthropometric markers. Overall, the results reveal an obesity prevalence of 23.23 %, 38.90 %, and 81.53 % according to BMI, WC, and WHtR, respectively, with increasing trends disproportionately affecting women. The review also highlighted significant variations in measurement methodologies and cut-off points used to determine obesity, leading to discrepancies in reported prevalences. Furthermore, the insufficiency of recent and specific data on obesity, especially in national studies, limits the ability to effectively track and adapt public health policies in response to this epidemic.
• •It is recommended that the collection and frequency of data on obesity in Peru be improved by implementing annual national surveys specifically designed for this purpose. These surveys should include a variety of anthropometric markers, such as weight, height, and WC, to obtain a more accurate and comprehensive representation of the epidemic. Additionally, it is crucial to standardize measurement methodologies and review the cut-off points used to adapt them to the specific characteristics of the Peruvian population, thereby improving the accuracy of prevalence estimates and allowing for more effective comparisons over time and between different studies.
• •From a public health policy perspective, emphasis should be placed on developing strategies that not only focus on reducing the prevalence of obesity but also address its root causes. This includes implementing educational programs on nutrition, promoting physical activity, and ensuring access to healthy foods. Policies and programs should be inclusive, considering gender and regional differences, and culturally appropriate to ensure effectiveness. Moreover, continuous research should be encouraged to monitor trends and evaluate the effectiveness of implemented interventions.

Informed consent

This study is a systematic review therefore informed consent is not required.

Author contributions

Luisa Erika Milagros Vásquez-Romero: Conceptualization, Investigation, Methodology, Resources, Writing - Original Draft, Writing - Review & Editing.

Fiorella E. Zuzunaga-Montoya: Investigation, Project administration, Writing - Original Draft, Writing - Review & Editing.

Joan A. Loayza-Castro: Investigation, Resources, Writing - Original Draft, Writing - Review & Editing.

Enrique Vigil-Ventura: Validation, Visualization, Writing - Original Draft, Writing - Review & Editing.

Willy Ramos: Software, Data Curation, Formal analysis, Writing - Review & Editing.

Víctor Juan Vera-Ponce: Methodology, Supervision, Funding acquisition, Writing - Review & Editing.

Ethical disclosure

Clinical Trial number: not applicable.

Human Ethics and Consent to Participate declarations: not applicable.

Data availability

Data are available upon request from the corresponding author.

Artificial intelligence disclosure

This manuscript was entirely written by the authors without the use of artificial intelligence tools. All analyses, interpretations, and conclusions are the sole work of the research team.

Financial disclosure

This study was financed by Vicerectorado de Investigación de la Universidad Nacional Toribio Rodríguez de Mendoza de Amazonas.

Conflict of interest

The authors declare no conflict of interest.

Appendix A. Supplementary data

The following is the Supplementary data to this article.